JP2592207Y2 - MRI equipment - Google Patents

Description

[Detailed description of the invention]

[0001]

This invention relates to an MRI apparatus,
More specifically, an MRI for extracting an electrocardiographic signal by removing a gradient magnetic field noise from an electrocardiographic signal on which a gradient magnetic field noise is superimposed by a gradient magnetic field noise removing filter using an adaptive filter.
Related to the device.

[0002]

2. Description of the Related Art FIG. 5 is an overall configuration diagram showing an example of a conventional MRI apparatus. In the MRI apparatus 51, the computer 2 controls the entire operation based on an instruction from the console 13. The sequence controller 3 activates the magnetic field drive circuit 4 based on the stored sequence, and generates a gradient magnetic field with the gradient magnetic field coil of the magnet assembly 5. Further, the gate modulation circuit 7 is controlled so as to modulate the RF signal generated by the RF transmission circuit 6, and the modulated R
The F signal is applied to the transmission coil of the magnet assembly 5 via the RF power amplifier 8. Magnet assembly 5
The NMR signal obtained by the receiving coil is input to the phase detector 10 via the preamplifier 9 and further input to the computer 2 via the A / D converter 11. Calculator 2 uses A
Based on the NMR signal data obtained from the / D converter 11,
The image data is calculated, and the image is displayed on the display device 12.

An electrode 18 and an ECG amplifier 19 collect an electrocardiographic signal from a subject. However, a gradient magnetic field noise is superimposed on the collected electrocardiographic signal. The electrocardiographic signal on which the gradient magnetic field noise is superimposed is converted into a digital value by the A / D converter 21 and input to the DSP 52.
The DSP 52 removes the gradient magnetic field noise from the electrocardiographic signal on which the gradient magnetic field noise is superimposed, extracts the electrocardiographic signal, and controls the scan execution timing based on the electrocardiographic signal.

FIG. 6 is an internal configuration diagram of the DSP 52.
The DSP 52 inputs an electrocardiographic signal d [i] (i; i) from the A / D converter 21 on which gradient magnetic field noise is superimposed.
From the time) to remove the gradient magnetic field noise and obtain an electrocardiographic signal e [i].
And a R-wave detector 42 that detects an R-wave from the electrocardiogram signal e [i] and outputs a scan execution timing signal Ra based on the R-wave to the sequence controller 3. Consists of

The gradient magnetic field noise removing filter 61 includes a filter coefficient adjusting unit 32, a gradient magnetic field noise reproducing unit 33,
And a gradient magnetic field noise removing unit 34. The filter coefficient adjusting unit 32 calculates a filter coefficient Bk [i] from the output of the gradient magnetic field noise removing unit 34, the gradient output waveform x [i] read from the magnetic field drive circuit 4, and a predetermined adaptive coefficient β1. . The gradient magnetic field noise reproducing unit 33 is an adaptive filter that reproduces the gradient magnetic field noise y [i] based on the filter coefficient B [i] input from the filter coefficient adjusting unit 32. The gradient magnetic field noise removing unit 34 removes the gradient magnetic field noise by subtracting the gradient magnetic field noise y [i] reproduced by the gradient magnetic field noise reproducing unit 33 from the electrocardiogram signal d [i] on which the gradient magnetic field noise is superimposed. And the ECG signal e
[I] is extracted.

FIG. 7 is a flowchart showing the operation of the gradient magnetic field noise elimination filter 61. Note that the gradient magnetic field noise reproducing unit 33 is an FIR filter, and the filter coefficient adjusting unit 32 adjusts the filter coefficient by the LSM algorithm. In step ST1, an initialization process is performed. In step ST2, the process waits until the electrocardiogram signal d [i] on which the gradient magnetic field noise is superimposed is input to the gradient magnetic field noise removing unit 34. In step ST3, the filter coefficient adjustment unit 32 reads the gradient output waveform x [i]. In step ST4, the gradient magnetic field noise reproducing unit 33 performs the next operation (Bk: filter coefficient, N: number of taps of the filter) to reproduce the gradient magnetic field noise y [i]. In step ST5, the gradient magnetic field noise elimination unit 34 performs the following calculation e [i] = d [i] -y [i] to extract the electrocardiographic signal e [i]. In step ST6, the gradient magnetic field noise removing unit 34 outputs the electrocardiographic signal e [i] to the R-wave detecting unit 42.

In step SU7, the filter coefficient adjustment unit 32 performs the following operation: Bk [i + 1] = Bk [i] + β1 · e [i] · x [ik] (β1; adaptive coefficient) Bk [i] is updated.

The processing in steps ST2 to SU7 is performed for each sampling cycle for reading the electrocardiographic signal d [i] on which the gradient magnetic field noise is superimposed.

[0009]

In the above-mentioned conventional MRI apparatus 51, if the adaptation coefficient β1 used in the gradient magnetic field noise removing filter 61 is increased, the adaptation speed is increased, so that the gradient magnetic field noise can be removed quickly from the start of scanning. However, gradually, not only the gradient magnetic field noise but also the electrocardiogram signal is removed. For this reason, conventionally, a slightly smaller adaptive coefficient β1 is set. However, there is a problem in that the adaptation speed is sacrificed, and in particular, the effect of removing the gradient magnetic field noise immediately after the start of scanning becomes insufficient. Therefore, an object of the present invention is to provide an MRI apparatus having an improved gradient magnetic field noise elimination filter so that gradient magnetic field noise can be quickly removed from the start of scanning and an electrocardiographic signal is not removed. is there.

[0010]

The MRI apparatus of the present invention extracts an electrocardiographic signal by removing gradient magnetic field noise from an electrocardiographic signal on which gradient magnetic field noise is superimposed by a gradient magnetic field noise removing filter using an adaptive filter. In the MRI apparatus, the adaptive coefficient β of the adaptive filter is set to a relatively large adaptive coefficient Lβ until a predetermined time elapses from the start of scanning, and thereafter the adaptive coefficient β of the adaptive filter is set to a relatively small adaptive coefficient Sβ The configuration is characterized by having the switching means.

[0011]

In the MRI apparatus according to the present invention, the adaptive coefficient β of the gradient magnetic field noise removing filter for removing noise from the electrocardiographic signal is set to a relatively large adaptive coefficient Lβ until a predetermined time has elapsed from the start of scanning. Thereafter, it is set to a relatively small adaptation coefficient Sβ. Therefore, immediately after the start of the scan, the adaptation speed of the gradient magnetic field noise removal filter increases, so that the gradient magnetic field noise can be quickly removed. After that, since the adaptation speed of the gradient magnetic field noise elimination filter is reduced, the electrocardiographic signal is not eliminated.

[0012]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. Note that this does not limit the present invention. FIG. 1 is an overall configuration diagram showing an embodiment of the MRI apparatus of the present invention. In the MRI apparatus 1, the computer 2 controls the entire operation based on an instruction from the console 13. The sequence controller 3 activates the magnetic field drive circuit 4 based on the stored sequence, and generates a gradient magnetic field with the gradient magnetic field coil of the magnet assembly 5. Further, the gate modulation circuit 7 is controlled so as to modulate the RF signal generated by the RF transmission circuit 6,
The modulated RF signal is applied to the transmission coil of the magnet assembly 5 via the RF power amplifier 8. The NMR signal obtained by the receiving coil of the magnet assembly 5 is input to the phase detector 10 via the preamplifier 9 and
The data is input to the computer 2 via the / D converter 11. The computer 2 calculates image data based on the data of the NMR signal obtained from the A / D converter 11 and displays the image on the display device 12.

The electrode 18 and the ECG amplifier 19 collect an electrocardiographic signal from the subject. However, a gradient magnetic field noise is superimposed on the collected electrocardiographic signal. The electrocardiographic signal on which the gradient magnetic field noise is superimposed is converted into a digital value by the A / D converter 21 and input to the DSP 22.
The DSP 22 removes the gradient magnetic field noise from the electrocardiographic signal on which the gradient magnetic field noise is superimposed, extracts an electrocardiographic signal, and controls the scan execution timing based on the electrocardiographic signal.

FIG. 2 is an internal block diagram of the DSP 22.
The DSP 22 inputs an electrocardiographic signal d [i] (i; i) input from the A / D converter 21 on which gradient magnetic field noise is superimposed.
From the time) to remove the gradient magnetic field noise and obtain an electrocardiographic signal e [i].
And a R-wave detector 42 that detects an R-wave from the electrocardiogram signal e [i] and outputs a scan execution timing signal Ra based on the R-wave to the sequence controller 3. Consists of

The gradient magnetic field noise removing filter 41 includes an adaptive coefficient switching unit 31, a filter coefficient adjusting unit 32, a gradient magnetic field noise reproducing unit 33, and a gradient magnetic field noise removing unit 34. The adaptive coefficient switching unit 31 inputs a relatively large adaptive coefficient Lβ to the filter coefficient adjusting unit 32 until a predetermined time elapses from the start of scanning, and thereafter switches to a relatively small adaptive coefficient Sβ and inputs it to the filter coefficient adjusting unit 32. .
The filter coefficient adjustment unit 32 includes a gradient magnetic field noise removal unit 34
, The gradient output waveform x [i] read from the magnetic field drive circuit 4 and the adaptive coefficient Lβ or Sβ input from the adaptive coefficient switching unit 31 to calculate a filter coefficient Bk [i]. The gradient magnetic field noise reproducing unit 33 is an adaptive filter that reproduces the gradient magnetic field noise y [i] based on the filter coefficient B [i] input from the filter coefficient adjusting unit 32. The gradient magnetic field noise removing unit 34 removes the gradient magnetic field noise by subtracting the gradient magnetic field noise y [i] reproduced by the gradient magnetic field noise reproducing unit 33 from the electrocardiogram signal d [i] on which the gradient magnetic field noise is superimposed. And the ECG signal e
[I] is extracted.

FIG. 3 is a flowchart showing the operation of the filter coefficient adjusting unit 32, the gradient magnetic field noise reproducing unit 33, and the gradient magnetic field noise removing unit 34 in the gradient magnetic field noise removing filter 41. Note that the gradient magnetic field noise reproducing unit 33 is an FIR filter, and the filter coefficient adjusting unit 32 is an LSM filter.
It is assumed that the filter coefficients are adjusted by an algorithm. In step ST1, an initialization process is performed. In step ST2, the process waits until the electrocardiogram signal d [i] on which the gradient magnetic field noise is superimposed is input to the gradient magnetic field noise removing unit 34. In step ST3, the filter coefficient adjustment unit 32 reads the gradient output waveform x [i]. In step ST4, the gradient magnetic field noise reproducing unit 33 performs the next operation (Bk: filter coefficient, N: number of filter taps) to reproduce the gradient magnetic field noise y [i]. In step ST5, the gradient magnetic field noise elimination unit 34 performs the following calculation e [i] = d [i] -y [i] to extract the electrocardiographic signal e [i]. In step ST6, the gradient magnetic field noise removing unit 34 outputs the electrocardiographic signal e [i] to the R-wave detecting unit 42.

In step ST7, the filter coefficient adjusting unit 32 calculates the following operation: Bk [i + 1] = Bk [i] + β ・ e [i] ・ x [ik] (β: adaptive coefficient, Lβ or Sβ Is performed, and the filter coefficient Bk [i] is updated.

The processing in steps ST2 to ST7 is performed for each sampling cycle for reading the electrocardiographic signal d [i] on which the gradient magnetic field noise is superimposed.

FIG. 4 is a flowchart showing the operation of the adaptive coefficient switching unit 31 in the gradient magnetic field noise elimination filter 41. In step SV1, a scan start notification J is received from the sequence controller 3 or the time is up by the built-in timer (time T from scan start> predetermined time Ts).
Wait until is notified. When the scan start notification is received, the process proceeds to step SV2, and when the time up is notified, the process proceeds to step SV4.

In step SV2, a relatively large adaptive coefficient Lβ is output as the adaptive coefficient β. In step SV3, the built-in timer is started, and measurement of the time T from the start of scanning is started. In step SV4, the adaptive coefficient β is switched to the relatively small adaptive coefficient Sβ and output. In step SV5, the internal timer is stopped.

In the above embodiment, the adaptive coefficient Lβ is relatively large during the period from the start of scanning until the time is up.
And the adaptive coefficient Sβ is relatively small when the time is up
However, three or more types of adaptive coefficients may be selectable, and may be switched to smaller adaptive coefficients as the time from the start of scanning becomes longer.

[0022]

According to the MRI apparatus of the present invention, the adaptation speed of the gradient magnetic field noise elimination filter using the adaptive filter increases immediately after the start of scanning and decreases after a predetermined time has elapsed. Therefore, immediately after the scan starts,
Gradient magnetic field noise can be quickly removed. After that, although the gradient magnetic field noise is removed, even the electrocardiographic signal is not removed. Therefore, it is suitable for performing an electrocardiographic gate scan and performing a cine scan.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram showing an embodiment of the MRI apparatus of the present invention.

FIG. 2 is an internal configuration diagram of a DSP of the MRI apparatus of FIG. 1;

FIG. 3 is a flowchart showing an operation of a gradient magnetic field noise elimination filter of the MRI apparatus of FIG. 1;

FIG. 4 is a flowchart showing the operation of an adaptive coefficient switching unit.

FIG. 5 is an overall configuration diagram showing an example of a conventional MRI apparatus.

6 is an internal configuration diagram of a DSP of the MRI apparatus of FIG.

FIG. 7 is a flowchart showing an operation of a gradient magnetic field noise elimination filter of the MRI apparatus of FIG. 5;

[Explanation of symbols]

 Reference Signs List 1,51 MRI apparatus 2 Computer 3 Sequence controller 4 Magnetic field drive circuit 5 Magnet assembly 13 Operation console 18 Electrode 19 ECG amplifier 21 A / D converter 22, 52 DSP 41, 61 Gradient magnetic field noise removal filter 31 Adaptive coefficient switching unit 32 Filter coefficient Adjustment unit 33 Gradient magnetic field noise reproducing unit 34 Gradient magnetic field noise removing unit

Claims (1)

(57) [Scope of request for utility model registration] 1. An MRI for extracting an electrocardiographic signal by removing a gradient magnetic field noise from an electrocardiographic signal on which the gradient magnetic field noise is superimposed by a gradient magnetic field noise removing filter using an adaptive filter.
The apparatus sets the adaptive coefficient β of the adaptive filter to a relatively large adaptive coefficient Lβ until a predetermined time elapses from the start of scanning, and thereafter sets the adaptive coefficient β of the adaptive filter to a relatively small adaptive coefficient Sβ. An MRI apparatus characterized by comprising means.